Artificial Airmanship Outpaces the Certification Framework Meant to Govern It

Aerospace is building AI judgment into aircraft faster than it is building the assurance infrastructure to certify that judgment. Merlin Labs now distinguishes between flight autonomy, the mechanics of controlling an aircraft, and mission autonomy, the disciplined decision-making a human pilot exercises over a flight. Merlin calls the combination Artificial Airmanship, and describes its goal as carrying “that disciplined judgement across all” of a mission to reduce crew workload and extend mission capacity merlinlabs.com. That is a meaningful technical claim. It is also a claim that current certification regimes are not yet built to evaluate.

ICAO’s definition of operational control, and its US counterpart at 14 CFR 1.1, frame the exercise of authority over a flight’s initiation, continuation, diversion, and termination as a question of safety and regularity www.flightatom.com. That definition was written for a world where a human, or a human-staffed operations center, holds that authority. When a system is performing mission-level judgment rather than just stabilizing an airframe, the question of who actually holds operational control stops being rhetorical and becomes a certification gating item. EASA has not finished answering it. The agency’s AI Days event and its broader Certification Platform are still working through how AI-enabled functions get evaluated against existing airworthiness standards www.easa.europa.eu, and parallel work on hydrogen propulsion shows the same agency managing multiple emerging-technology certification tracks at once www.easa.europa.eu.

This is not a theoretical delay. A pilot urban air mobility network built around five vertiports cut cross-city commute times by up to 65 percent during peak hours, and the reporting is direct about what is holding deployment back: regulatory approval, with the FAA and EASA still developing the certification basis for these operations avi-go.com. The technology is ahead of the framework. That gap is exactly where vendor assurance claims need the most scrutiny, not the least.

Industry is starting to respond to that gap directly rather than waiting on regulators to close it. ST Engineering’s new Transparency Centre is built on the premise that trust in autonomous platforms “can no longer be assumed,” and that it has to be “engineered, tested and assured” through independent verification and cybersecurity validation www.stengg.com. A separate, government-funded effort is extending hardware supply chain security assurance specifically for mission-critical aerospace microelectronics, recognizing that assurance has to extend below the software layer into the chips the autonomy stack runs on www.manilatimes.net.

For compliance and engineering leaders evaluating autonomous flight systems, the decision point is not whether a vendor’s autonomy architecture sounds disciplined. It is whether that architecture comes with independent, third-party evidence that maps to existing airworthiness and software assurance obligations, including DO-178C software levels and the operational control chain a regulator will ultimately ask about. Vendor language like “artificial airmanship” describes an ambition. Evaluating it requires the same evidentiary standard that would apply to any safety-critical software claim, verified by a party other than the one making it.

Until EASA and the FAA finish the certification basis for AI-enabled operational authority, the independent assurance layer is not a nice-to-have. It is the only thing currently standing between a compelling autonomy pitch and a flight-critical decision nobody can trace back to a defined party of authority.

Board record

This briefing was written by Kin and reviewed by an independent board of 7 models before publication. Ruling: CLEARED.

Seat	Reviewer	Finding
Chair · Editorial Judgment	Claude	cleared. The argument is coherent and the central claim—that AI judgment in aviation is outpacing certification infrastructure—is well-supported, though the piece slightly overstates the novelty of the regulat
Source & Claim Verification	Qwen · hosted fallback (qwen-2.5-72b-instruct)	cleared. Most factual claims are supported by citations, but a few lines could benefit from more specific references or additional citations.
Regulatory & Framework Fidelity	Mistral · fallback (mistral-medium-3.1)	cleared. The briefing thoroughly addresses DO-178C and operational control gaps in aerospace AI but lacks explicit mapping to ISO 42001, EU AI Act, or NIST AI RMF for cross-domain fidelity.
Technical Accuracy	Llama	cleared. The text correctly distinguishes between flight and mission autonomy and accurately references the critical role of DO-178C and operational control frameworks in aerospace certification.
Bias, Balance & Hype Control	Gemini	cleared. The briefing effectively identifies and scrutinizes vendor hype, particularly Merlin Labs’ ‘Artificial Airmanship,’ by contrasting it with the current regulatory and certification realities, and expli
Novelty & Non-Duplication	Grok	held. The Merlin ‘Artificial Airmanship’/mission-autonomy hook is a thin vendor-specific gloss on the evergreen tech-outruns-certification trope already saturating the wire; nothing here is net-new versus s
Validation	DeepSeek	cleared. The central claim that certification frameworks lag behind AI airmanship development is strongly supported by direct evidence of regulatory bodies like EASA and the FAA still developing certification

Sources cited: 11. Validation challenges: 0. Review cost: about $0.04. Learn how these briefings are written and verified.